Our patient who had a steroid-resistant exacerbation of bronchial asthma during mepolizumab therapy was diagnosed with PCP based on ground-glass opacities on computed tomography and positive polymerase chain reaction test results for Pneumocystis DNA in the sputum. The patient was treated with SMX–TMP, which improved the symptoms. Therefore, PCP was considered as a possible cause of bronchial asthma exacerbation during anti-IL-5 antibody therapy.
There are two types of PCP: HIV-associated PCP, which occurs in patients with immunosuppression due to HIV infection, and non-HIV-associated PCP, which develops independently of HIV infection. Non-HIV-associated PCP has an acute course and is more severe than HIV-associated PCP. Moreover, the amount of pathogen at the time of infection is low and difficult to detect on speculum examination [8]. The pathogenesis of non-HIV-associated PCP is believed to be associated with an excessive immune response, and is correlated with inflammatory cytokines such as IL-4, IL-5, tumor necrosis factor-α, and interferon-γ [9, 10]. In mice, loss of STAT6, a transcription factor involved in Th2 cell differentiation, suppresses bronchial hypersensitivity [9]. Since Th2-type inflammatory responses are strongly associated with airway hypersensitivity symptoms in non-HIV-associated PCP, Pneumocystis spp. infection might have contributed to exacerbation and refractoriness of asthma symptoms in this case.
Pneumocystis spp. carriage rates range from 16–55% and 0–65% in patients with COPD and patients receiving immunosuppressive therapy, respectively [11]. Although there is no report on the rate of carriage in people with asthma, it is highly likely that our patient was carrying the disease, since anti-Pneumocystis antibodies are higher in patients with severe asthma and in older patients [7]. In this patient, PCP developed in the absence of any background HIV infection or disease indicating compromised immunity. In general, PSL 16 mg/day for > 8 weeks increases the risk of developing PCP [12], and prophylactic administration of SMX–TMP is recommended when PSL 20 mg/day is administered for ≥ 4 weeks [13]. In the present case, the patient had received < 5 mg/day of PSL previously and 10 mg/day of PSL for 2 weeks prior to admission. Therefore, no prophylactic SMX–TMP was administered because the dose and duration of administration of PSL did not indicate the need for prophylactic SMX–TMP administration. Thus, other factors were possibly involved in the pathogenesis of Pneumocystis pneumonia.
Another factor in the development of Pneumocystis pneumonia is the use of anti-IL-5 antibodies. Eosinophils have immunomodulatory functions in allergic diseases as well as in bacterial, fungal, and viral infections [14], and have been implicated in immunity against Pneumocystis spp. [15]. They have been shown to have killing activity against Pneumocystis spp. in a mouse model [15], suggesting that immunity against Pneumocystis jiroveci might be mediated by eosinophils in humans as well. In this case, the decrease in eosinophils caused by the use of anti-IL5 antibody was possibly responsible for the development of Pneumocystis pneumonia.
In conclusion, Pneumocystis sp. infection should be considered as a differential diagnosis in patients with severe asthma that has been well controlled with biologic agents who develop refractory exacerbations.